Since it is not possible to encounter a smell in our external environment without first inhaling through the nose, stimulus sampling in the olfactory system is inextricably linked to breathing. Respiratory-driven local field potential (LFP) oscillations are important for odor coding mechanisms in the rodent olfactory bulb, but their role in higher olfactory structures such as piriform cortex is not well understood, with even less known about respiratory oscillations in the human brain. While breathing drives oscillations in the brain, the reverse must also be true; stimulus sampling in the olfactory system requires overriding of autonomic respiratory rhythms in order to achieve intentional sniffing and fast adaptive sniff modifications in response to chemical stimuli. The overarching goal of this proposal is to understand the function of respiratory oscillations in the human brain, including their role in the formation of odor-evoked responses in olfactory brain regions and fostering communication across limbic networks involved in odor sampling and fast adaptive sniffing modifications. We also aim to elucidate limbic networks involved in olfactory sampling behaviors. To measure LFPs from medial olfactory structures in the human brain, we will use intracranial electroencephalography (iEEG) with a high sampling rate (up to 10,000Hz), allowing analysis of limbic LFP oscillations across a range of frequencies. We will use a combination of iEEG, direct electrical stimulation, psychophysics and functional neuroimaging and tractography techniques to accomplish the goals of three Specific Aims. First, we will test the hypothesis that slow respiratory-linked LFP oscillations organize the spectral and temporal structure of odor-evoked responses in human piriform cortex. To isolate the impact of slow respiratory-driven oscillations on odor codes, we will deliver odors in the presence and absence of sniffs, accomplished by velopharyngeal closure paired with artificial air flow through the nose. Second, we will test the hypothesis that slow respiratory oscillations across a limbic network of regions important for respiratory control mediate odor sampling, or sniffing behaviors. Here we will use iEEG techniques, electrical stimulation and MRI techniques to study limbic networks involved in the control of nasal breathing with a particular emphasis on the amygdala. Third, we will use iEEG, electrical stimulation and psychophysics to test the hypothesis that adaptive fast sniffing reductions in response to potentially threatening odors are mediated by the amygdala, and can generalize to non-olfactory stimuli in anxious states. The proposed studies have several direct clinical applications. Research on Sudden Unexpected Death in Epilepsy (SUDEP), the most common cause of death in patients with Epilepsy, implicates respiratory dysfunction as a potential cause, with converging evidence for an amygdalar role in the disease (13,14). In so far as our proposal aims to deepen understanding of the neural mechanisms of the amygdala's role in respiratory control, these studies will be important in gaining a better understanding of SUDEP. Our research will also elucidate dysfunctional olfactory-limbic networks underlying clinical anxiety.